文档介绍：Computational Economics (2005) 26: 65–89
DOI: -005-7366-2 C Springer 2005
Detecting Business Cycle Asymmetries Using
Artificial works and Time Series Models
KHURSHID M. KIANI
Kansas State University, Kansas, .; E-mail:
Accepted 17 May 2005
Abstract. This study examines possible existence of business cycle asymmetries in Canada, France,
Japan, UK, and USA real GDP growth rates using works nonlinearity tests and tests based
on a number of nonlinear time series models. These tests are constructed using in-sample forecasts
from artificial works (ANN) as well as time series models.
Our study results based on work tests show that there is statistically significant evidence
of business cycle asymmetries in these industrialized countries. Similarly, our study results based on a
number of time series models also show that business cycle asymmetries do prevail in these countries.
So we are not able to evaluate the impact of ary policy or any other shocks on GDP in these
countries based on linear models.
Keywords: B22, C32, C45, E32
1. Introduction
People were not familiar with business cycles until eenth century when
economists thought of using new economic tools to learn the consequences of shifts
in aggregate supply and demand. From early twentieth century, economists started
thinking of business cycles and underlying factors affecting them. Indeed, the urge
to understand business cycles intensified after the great depression of 1930s. Key-
nesian macroeconomics remained in tact after 1930s great depression. However,
following Lucas (1976), Kydland and Prescott (1982) and Long and Plosser (1983)
business cycle models do not pass ary factors and money management.
Schumpeter (1939) classified business cycles into a number of types according to
their peak-to-peak and trough-to-trough durations. He also divided business cycles
into four phases, . recession, depression, recovery and boom. Many macroeco-
nomic variables follow thes